The incidence of non-Hodgkin's lymphoma has been increasing at a rate of 4% per year since 1950; more than 62,000 cases will be diagnosed in the United States in 2000. Diffuse large cell lymphoma (DLCL) is the prototype of curable non-Hodgkin's lymphoma. Empirically designed chemotherapy regimens did not increase the cure rate of 30-40% achieved by the original four-drug regimen introduced in the 1970s [cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP)]. We studied the antitumor effects of the CHOP regimen alone or in combination with a unique protein kinase C activator, bryostatin 1, on a xenograft model for resistant DLCL in mice with severe combined immune deficiency (WSU-DLCL2-SCID). In this model, the efficacy of bryostatin 1 given at 75 microg/kg, i.p., alone for 1 or 2 days [B(1x) and B(2x)]was compared with the efficacy of CHOP alone, bryostatin 1 + CHOP (B+CHOP) given concurrently, bryostatin 1 for 1 day followed by CHOP on day 2 [B(1x)-CHOP], and bryostatin 1 for 2 days followed by CHOP on day 3 [B(2x)-CHOP]. CHOP doses were as follows: (a) cyclophosphamide, 40 mg/kg, i.v.; (b) doxorubicin, 3.3 mg/kg, i.v.; (c) vincristine, 0.5 mg/kg, i.v.; and (d) prednisone, 0.2 mg/kg, every day for 5 days, p.o. Tumor growth inhibition (T/C), tumor growth delay (T-C), and log10 kill for B(1x), B(2x), CHOP, B+CHOP, B(1x)-CHOP and B(2x)-CHOP were 49%, 39%, 25.8%, 15.1%, 14.6%, and 12%; 6, 7, 16, 25, 12, and 15 days; and 0.6, 0.5, 2.2, 3.6, 1.7, and 2.0, respectively. To begin elucidating the mechanism whereby bryostatin 1 potentiated the effects of CHOP in the mouse model; we studied the effect of bryostatin 1 on Bax, Bcl-2, and poly(ADP-ribose) polymerase proteins in vitro and in vivo. Bax protein increased in a time-dependent manner without any measurable change in Bcl-2 expression. However, significant cleavage of the preapoptotic marker poly(ADP-ribose) polymerase was not recorded, and the percentage of apoptotic cells detected by flow cytometry increased only slightly (approximately 8%) after 96 h of bryostatin 1 exposure. The in vitro and in vivo results emphasize the superiority of combining bryostatin 1 with the CHOP regimen against the WSU-DLCL2 model. One possible mechanism may be the modulatory effects of bryostatin 1 on the Bax:Bcl-2 family of apoptosis-regulatory proteins. The use of this combination should be further explored clinically in the treatment of lymphoma.